README

GitHub Documents

TEST2 This is an R Markdown format used for publishing markdown documents to GitHub. When you click the Knit button all R code chunks are run and a markdown file (.md) suitable for publishing to GitHub is generated.

데이터 불러오기

S-Dot 1시간 측정 평균값이 저장된 csv파일 read

#분석하고자 하는 csv데이터를 sdot이라는 데이터 프레임에 넣기 
sdot<-read.csv("data/sdot_20200507.csv", fileEncoding = 'euc-kr', head=TRUE, check.names=FALSE) 

#데이터 중 위에서부터 100개의 데이터만 확인
paged_table(sdot) %>% head(100)

sdot에 저장된 값에 대한 데이터 요약

#명령어 summary(데이터프레임)
summary(sdot) 

##          시리얼            date            hour       초미세먼지 (㎍/㎥)
##  V02Q1940043:   24   2020.5.1:20280   Min.   : 0.00   Min.   :  0.00    
##  V02Q1940044:   24                    1st Qu.: 5.75   1st Qu.: 14.10    
##  V02Q1940045:   24                    Median :11.50   Median : 23.69    
##  V02Q1940046:   24                    Mean   :11.50   Mean   : 23.56    
##  V02Q1940047:   24                    3rd Qu.:17.25   3rd Qu.: 31.00    
##  V02Q1940049:   24                    Max.   :23.00   Max.   :175.17    
##  (Other)    :20136                                                      
##  미세먼지 (㎍/㎥)   기온 (℃)      상대습도 (%)     풍향 (°)      
##  Min.   :  0.00   Min.   :15.24   Min.   :31.57   Min.   :  0.000  
##  1st Qu.: 18.82   1st Qu.:19.46   1st Qu.:57.06   1st Qu.:  0.000  
##  Median : 31.90   Median :21.21   Median :68.27   Median :  0.000  
##  Mean   : 31.75   Mean   :22.02   Mean   :64.78   Mean   :  2.363  
##  3rd Qu.: 41.67   3rd Qu.:24.41   3rd Qu.:72.12   3rd Qu.:  0.000  
##  Max.   :268.33   Max.   :34.23   Max.   :85.62   Max.   :299.593  
##                                                                    
##    풍속 (m/s)      돌풍 풍향 (°)    돌풍 풍속 (m/s)     조도 (lux)      
##  Min.   :0.00000   Min.   :  0.000   Min.   :0.00000   Min.   :    1.00  
##  1st Qu.:0.00000   1st Qu.:  0.000   1st Qu.:0.00000   1st Qu.:    5.12  
##  Median :0.00000   Median :  0.000   Median :0.00000   Median : 1296.71  
##  Mean   :0.01573   Mean   :  2.328   Mean   :0.03093   Mean   : 6664.73  
##  3rd Qu.:0.00000   3rd Qu.:  0.000   3rd Qu.:0.00000   3rd Qu.: 9079.19  
##  Max.   :4.55714   Max.   :295.750   Max.   :8.40714   Max.   :50200.48  
##                                                                          
##   자외선 (UVI)       소음 (dB)      진동(x) (g)       진동(y) (g)     
##  Min.   :  1.000   Min.   : 0.00   Min.   :0.00000   Min.   :0.00000  
##  1st Qu.:  2.000   1st Qu.:45.38   1st Qu.:0.01833   1st Qu.:0.02815  
##  Median :  2.130   Median :48.07   Median :0.02276   Median :0.04826  
##  Mean   :  9.105   Mean   :48.17   Mean   :0.02797   Mean   :0.05542  
##  3rd Qu.: 10.823   3rd Qu.:50.90   3rd Qu.:0.03500   3rd Qu.:0.07880  
##  Max.   :108.767   Max.   :64.52   Max.   :0.11077   Max.   :0.17696  
##                                                                       
##   진동(z) (g)    진동(x) 최대 (g)  진동(y) 최대 (g) 진동(z) 최대 (g)
##  Min.   :0.000   Min.   :0.00000   Min.   :0.0000   Min.   : 0.000  
##  1st Qu.:1.006   1st Qu.:0.07241   1st Qu.:0.1163   1st Qu.: 1.095  
##  Median :1.029   Median :0.08708   Median :0.1300   Median : 1.107  
##  Mean   :1.026   Mean   :0.09167   Mean   :0.1340   Mean   : 1.117  
##  3rd Qu.:1.050   3rd Qu.:0.10088   3rd Qu.:0.1454   3rd Qu.: 1.124  
##  Max.   :1.116   Max.   :2.41815   Max.   :2.0540   Max.   :10.451  
##                                                                     
##  흑구 온도 (℃)    미세먼지 보정 (㎍/㎥) 초미세먼지 보정 (㎍/㎥)
##  Min.   :  4.833   Min.   :-88.12        Min.   :-98.12         
##  1st Qu.:  5.000   1st Qu.: 36.09        1st Qu.: 23.23         
##  Median :  5.000   Median : 46.00        Median : 27.76         
##  Mean   :  7.590   Mean   : 43.68        Mean   : 26.87         
##  3rd Qu.:  5.000   3rd Qu.: 53.31        3rd Qu.: 32.00         
##  Max.   :195.385   Max.   :345.60        Max.   :179.97         
## 

위의 값을 확인 했을때 현재 살펴보고자 하는 미세먼지, 초미세먼지 결측값, 음수, 0값을 제거

# 분석 편의상 컬럼 이름 변경 
sdot <- sdot %>% dplyr::rename(super_dust =`초미세먼지 보정 (㎍/㎥)`)
sdot <- sdot %>% dplyr::rename(dust =`미세먼지 보정 (㎍/㎥)`)
sdot <- sdot %>% dplyr::rename(model =시리얼)

# Na또는 Nan등 결측값 확인 
sum(is.na(sdot)) 

## [1] 0

0개

#미세먼지 값이 0이하 or 초미세먼지 값이 0 이하 값 제거
sdot <-sdot %>% filter(!super_dust <=0 , !dust <=0)


#명령어 summary(데이터프레임)로 미세먼지 보정, 초미세먼지 보정 값 재확인 
summary(sdot) 

##          model             date            hour       초미세먼지 (㎍/㎥) 
##  V02Q1940043:   24   2020.5.1:19960   Min.   : 0.00   Min.   :  0.03333  
##  V02Q1940044:   24                    1st Qu.: 5.00   1st Qu.: 14.37931  
##  V02Q1940047:   24                    Median :11.00   Median : 23.84413  
##  V02Q1940050:   24                    Mean   :11.49   Mean   : 23.81125  
##  V02Q1940052:   24                    3rd Qu.:18.00   3rd Qu.: 31.07407  
##  V02Q1940054:   24                    Max.   :23.00   Max.   :175.16667  
##  (Other)    :19816                                                       
##  미세먼지 (㎍/㎥)      기온 (℃)      상대습도 (%)     풍향 (°)      
##  Min.   :  0.06667   Min.   :15.24   Min.   :31.57   Min.   :  0.000  
##  1st Qu.: 19.28325   1st Qu.:19.46   1st Qu.:57.10   1st Qu.:  0.000  
##  Median : 32.08333   Median :21.21   Median :68.29   Median :  0.000  
##  Mean   : 32.08483   Mean   :22.02   Mean   :64.81   Mean   :  2.382  
##  3rd Qu.: 41.78571   3rd Qu.:24.41   3rd Qu.:72.13   3rd Qu.:  0.000  
##  Max.   :268.33333   Max.   :34.23   Max.   :85.62   Max.   :299.593  
##                                                                       
##    풍속 (m/s)      돌풍 풍향 (°)    돌풍 풍속 (m/s)     조도 (lux)   
##  Min.   :0.00000   Min.   :  0.000   Min.   :0.00000   Min.   :    1  
##  1st Qu.:0.00000   1st Qu.:  0.000   1st Qu.:0.00000   1st Qu.:    5  
##  Median :0.00000   Median :  0.000   Median :0.00000   Median : 1286  
##  Mean   :0.01595   Mean   :  2.349   Mean   :0.03131   Mean   : 6672  
##  3rd Qu.:0.00000   3rd Qu.:  0.000   3rd Qu.:0.00000   3rd Qu.: 9088  
##  Max.   :4.55714   Max.   :295.750   Max.   :8.40714   Max.   :50200  
##                                                                       
##   자외선 (UVI)       소음 (dB)      진동(x) (g)       진동(y) (g)     
##  Min.   :  1.000   Min.   : 0.00   Min.   :0.00000   Min.   :0.00000  
##  1st Qu.:  2.000   1st Qu.:45.39   1st Qu.:0.01833   1st Qu.:0.02800  
##  Median :  2.120   Median :48.04   Median :0.02267   Median :0.04821  
##  Mean   :  9.097   Mean   :48.15   Mean   :0.02786   Mean   :0.05543  
##  3rd Qu.: 10.828   3rd Qu.:50.88   3rd Qu.:0.03467   3rd Qu.:0.07900  
##  Max.   :108.767   Max.   :64.52   Max.   :0.11077   Max.   :0.17696  
##                                                                       
##   진동(z) (g)    진동(x) 최대 (g)  진동(y) 최대 (g) 진동(z) 최대 (g)
##  Min.   :0.000   Min.   :0.00000   Min.   :0.0000   Min.   : 0.000  
##  1st Qu.:1.006   1st Qu.:0.07241   1st Qu.:0.1163   1st Qu.: 1.095  
##  Median :1.029   Median :0.08700   Median :0.1300   Median : 1.107  
##  Mean   :1.026   Mean   :0.09149   Mean   :0.1340   Mean   : 1.117  
##  3rd Qu.:1.050   3rd Qu.:0.10067   3rd Qu.:0.1453   3rd Qu.: 1.124  
##  Max.   :1.116   Max.   :2.41815   Max.   :2.0540   Max.   :10.451  
##                                                                     
##  흑구 온도 (℃)         dust            super_dust       
##  Min.   :  4.833   Min.   :  0.1333   Min.   :  0.03448  
##  1st Qu.:  5.000   1st Qu.: 36.8667   1st Qu.: 23.48148  
##  Median :  5.000   Median : 46.2069   Median : 27.87750  
##  Mean   :  7.627   Mean   : 44.3847   Mean   : 27.40362  
##  3rd Qu.:  5.000   3rd Qu.: 53.4511   3rd Qu.: 32.07692  
##  Max.   :195.385   Max.   :345.6000   Max.   :179.96667  
## 

#결측값 및 0이하 값 제거 확인

#시간에 따른 미세먼지 값 확인 
#동적 그래프 생성규칙 

#plot_ly(데이터프레임, x = ~시간, y = ~값)
plot_ly(sdot, x = ~hour, y = ~dust, name = 'trace 0', type = 'scatter', mode = 'markers',
        marker = list(opacity = 0.4, color = "blue"))

plot_ly(sdot, x = ~hour, y = ~super_dust, name = 'trace 0', type = 'scatter', mode = 'markers',
        marker = list(opacity = 0.4, color = "red"))

#센서 설치정보 불러오기 
model <- read.csv("data/model.csv", head = TRUE, fileEncoding = 'euc-kr')

#SDot 데이터와 설치 정보 결합하기 
sdot_model <- merge(sdot, model, by ="model", all.x = TRUE)

#경도 위도 이름 변경 및 사용데이터 결합
sdot_model <- sdot_model %>% dplyr::rename(long = y, lat = x)
sdot_model <- sdot_model %>% group_by(gover, model, hour) %>% summarise(long = mean(long), lat = mean(lat), super_dust = mean(super_dust), dust = mean(dust))

## `summarise()` regrouping output by 'gover', 'model' (override with `.groups` argument)

지도 데이터 불러오기

지도정보 불러오기 :

#서울시/자치구/이름 파일 
p<- read.any("data/sample.csv", header = TRUE) #시각화할 데이터셋

#대한민국 시군구 행정구역 shp파일 
map =readOGR('data/TL_SCCO_SIG.shp',encoding = 'cp949')

## OGR data source with driver: ESRI Shapefile 
## Source: "/Users/gim-ilhong/Documents/mac/Github_S-Dot/S_DoT/data/TL_SCCO_SIG.shp", layer: "TL_SCCO_SIG"
## with 250 features
## It has 3 fields

#행정구역 shp 전처리
map <- spTransform(map, CRSobj = CRS('+proj=longlat +ellps=WGS84 +datum=WGS84 +no_defs'))
map@polygons[[1]]@Polygons[[1]]@coords %>% head(n = 10L)

##           [,1]     [,2]
##  [1,] 127.0086 37.58047
##  [2,] 127.0087 37.58045
##  [3,] 127.0088 37.58044
##  [4,] 127.0089 37.58042
##  [5,] 127.0091 37.58039
##  [6,] 127.0092 37.58039
##  [7,] 127.0092 37.58038
##  [8,] 127.0093 37.58038
##  [9,] 127.0093 37.58038
## [10,] 127.0094 37.58036

new_map <- fortify(map, region = 'SIG_CD')

## SpP is invalid

new_map$id <- as.numeric(new_map$id)

#대한민국 시군구 행정구역 중 서울시만 불러오기 
seoul_map <- new_map[new_map$id <= 11740,]

#서울시 자치구 이름 표기 정보 결합
P_merge <- merge(seoul_map, p, by='id')
P_merge <- P_merge %>% dplyr::rename(gover = 시군구명)

top_stat_sdot<-sdot_model %>% group_by(gover, hour) %>% summarise(dust = mean(dust), super_dust = mean(super_dust))

## `summarise()` regrouping output by 'gover' (override with `.groups` argument)

top_merge<-merge(P_merge, top_stat_sdot, by="gover", all.x =TRUE)


################### 아주 중요 ##############################
split_data_poly = lapply(unique(P_merge$group), function(x) {
  df = as.matrix(P_merge[P_merge$group == x, c("long", "lat")])
  polys = Polygons(list(Polygon(df)), ID = x)
  return(polys)
})

#split_data_poly
data_polys = SpatialPolygons(split_data_poly)


#지도에 SDOT 설치 정보 뿌리기
#테스트버전
leaflet(data_polys) %>%
  setView(lng=126.9784, lat=37.566, zoom=11) %>%
  #자치구 경계선 
  addPolygons(fillColor = "white",weight ="2",color = "black", opacity = 0.8   ) %>% 
  #맵 스타일 
  addProviderTiles('CartoDB.Positron') %>% 
  #sdot_model에 담겨있는 위치정보를 맵에 보여주기 
  addCircleMarkers(data = sdot_model, lng=~long, lat=~lat, color="#20639B", radius = "2",stroke = TRUE, fillOpacity = 0.8, weight =1)

# 미세먼지 값 표시하기
# 미세먼지 범례설정 
sdot_model$dust_range <- cut(sdot_model$dust, 
                             c(0,30,80,150,999), include.lowest = T,
                             labels = c('0','15','35','75'))
colors_sdot <- c("#3982BA", "#66B74F", "#FAE284", "#E93A30")
dust_color_sdot <- colorFactor(palette = colors_sdot , sdot_model$dust_range)


# 초미세먼지 범례설정
sdot_model$super_dust_range <- cut(sdot_model$super_dust, 
                             c(0,15,35,75,999), include.lowest = T,
                             labels = c('0','15','35','75'))
colors_sdot <- c("#3982BA", "#66B74F", "#FAE284", "#E93A30")
super_dust_color_sdot <- colorFactor(palette = colors_sdot , sdot_model$super_dust_range)




#범례 커스텀 옵션
addLegendCustom <- function(map, position, colors, labels, sizes, opacity = 0.5, title){
  colorAdditions <- paste0(colors, "; border-radius: 50%; width:", sizes, "px; height:", sizes, "px")
  labelAdditions <- paste0("<div style='display: inline-block;height: ", 
                           sizes, "px;margin-top: 4px;line-height: ", sizes, "px;'>", 
                           labels, "</div>")
  
  return(addLegend(map, position = "bottomright",
                   colors = colorAdditions, 
                   labels = labelAdditions, opacity = opacity, title = title))
}



# 일별 
mean_sdot_model <- sdot_model %>% group_by(model) %>% summarise(super_dust =mean(super_dust), dust = mean(dust), long = mean(long), lat = mean(lat))

## `summarise()` ungrouping output (override with `.groups` argument)

# 일별 미세먼지 범례설정 
mean_sdot_model$dust_range <- cut(mean_sdot_model$dust, 
                             c(0,30,80,150,999), include.lowest = T,
                             labels = c('0','15','35','75'))
colors_sdot <- c("#3982BA", "#66B74F", "#FAE284", "#E93A30")
dust_color_sdot <- colorFactor(palette = colors_sdot , mean_sdot_model$dust_range)


# 일별 초미세먼지 범례설정
mean_sdot_model$super_dust_range <- cut(mean_sdot_model$super_dust, 
                             c(0,15,35,75,999), include.lowest = T,
                             labels = c('0','15','35','75'))
colors_sdot <- c("#3982BA", "#66B74F", "#FAE284", "#E93A30")
super_dust_color_sdot <- colorFactor(palette = colors_sdot , mean_sdot_model$super_dust_range)


# 미세먼지 농도 지도에 표기
leaflet(data_polys) %>%
  setView(lng=126.9784, lat=37.566, zoom=11) %>%
  addPolygons(fillColor = "white",weight ="2",color = "black", opacity = 0.8  ) %>% 
  addProviderTiles('CartoDB.Positron') %>% 
  addCircleMarkers(data = mean_sdot_model, lat = ~lat, lng = ~long,
                   color = ~dust_color_sdot(dust_range), popup = sdot_model$dust,
                   radius = ~sqrt(dust/3),  stroke = FALSE, fillOpacity = 0.2, ) %>%
  addLegendCustom("bottomright",
                  colors =c("#3982BA", "#66B74F", "#FAE284", "#E93A30"),
                  labels= c("좋음 : 0 ~ 30㎍/㎥", "보통 : 31 ~ 80㎍/㎥","나쁨 : 81 ~ 150㎍/㎥","매우나쁨 : 151㎍/㎥ 이상"),
                  sizes = c(10, 12, 13,15),
                  opacity = 1,
                  title= "미세먼지(PM10) 기준 등급")